Systems and methods are presented that detect and classify mass-like regions exhibiting spiculated and/or dense characteristics with high sensitivity and at acceptable false positive rates. One or more suspicious masses are identified in medical imagery of the breast. In certain embodiments, a quantitative measure of spiculation and quantitative measure of density are computed for each suspicious mass located. At least one classification scheme, developed using true and false positives with similar quantitative measures, is then selected for each suspicious mass according to both quantitative measures. In certain other embodiments, a measure of breast location is computed for each suspicious mass. In one embodiment, the location determines whether a suspicious mass appears inside or outside of the parenchyma region of the breast.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A system for the automated classification of suspicious masses in medical imagery of the breast, the system comprising: a. means for locating one or more suspicious masses in medical imagery of the breast; b. means for computing, for one or more of the suspicious masses located, a quantitative measure of spiculation of the suspicious mass and a quantitative measure of density of the suspicious mass; c. means for selecting at least one classification scheme from a plurality of classification schemes for classifying the suspicious mass, wherein the selection is based on the quantitative measure of spiculation and the quantitative measure of density; and d. means for assigning the suspicious mass to a class by executing the selected at least one classification scheme on the suspicious mass.
2. A system for the automated classification of suspicious masses in medical imagery of the breast, the system comprising: a. means for detecting one or more candidate suspicious masses in medical imagery of the breast; b. means for characterizing the spiculatedness of each of the one or more candidate masses; c. means for characterizing the density of each of the one or more candidate masses; d. means for selecting at least one feature set from a plurality of predefined feature sets based at least on the characterized spiculatedness and the characterized density of each of the one or more candidate masses; and e. means for determining the suspiciousness of each of the one or more candidate masses based at least on the selected at least one feature set.
3. The system of claim 2 , wherein the medical imagery of the breast is a mammogram.
4. The system of claim 2 , wherein the means for selecting comprises a means for selecting at least one feature set based on whether the characterized spiculatedness exceeds a spiculatedness threshold and whether the characterized density exceeds a density threshold.
5. A system for the automated detection of cancers in a medical image, comprising: a. a detector for detecting one or more candidate cancers in the medical image; b. a characterizer for characterizing each of the detected one or more candidate cancers on the basis of a set comprising two or more predetermined characteristics known to distinguish cancers from non-cancers; c. a set of at least three software classifiers for determining the suspiciousness of the detected one or more candidate cancers, wherein each software classifier is designed to distinguish cancers that exhibit a unique combination of predetermined characteristics in the set from non-cancers that exhibit the unique combination of predetermined characteristics in the set.
6. The system of claim 5 , wherein the two or more predetermined characteristics include density and spiculatedness.
7. The system of claim 6 , wherein the set of at least three software classifiers consists of: at least one software classifier designed to distinguish cancers that exhibit only density characteristics from non-cancers that exhibit only density characteristics; at least one software classifier designed to distinguish cancers that exhibit only spiculatedness characteristics from non-cancers that exhibit only density characteristics; and at least one software classifier designed to distinguish cancers that exhibit density characteristic and spiculatedness characteristics from non-cancers that exhibit density characteristic and spiculatedness characteristics.
8. The system of claim 6 , wherein the detector detects candidate breast mass cancers in mammograms.
9. The system of claim 6 , further comprising a graphical user interface for outputting indicators as to one or more candidate masses determined to be suspicious.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
May 7, 2012
April 1, 2014
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.